In 1995 there were more than 450,000 cardiac surgery operations in the USA, of which 75 percent were coronary artery bypass graft (CABG) procedures. With these procedures increasing at a rate of 4 percent per year, there will be over 500,000 cardiac operations performed in 850 centers in 1999. Although in the past two decades mortality has declined as a result of improvements in cardiac surgery, the average age of today's patient is 10 years higher than in the 1970's, with a concomitant increase in morbidity. Up to 5 percent of cardiac operations result in patients with frank stroke and 50-80 percent of patients experience a measurable decrement in neurobehavioral function. A substantial amount of our earlier work was directed to defining sensitive state-of-the-art measures of neuropsychologic dysfunction and utilizing these tests to define the magnitude of this problem. However, little has been published about the etiology of these neurobehavioral deficits or their prevention. It is known that atherosclerotic disease (atheroma) of the ascending aorta and arch (atheroma) are predictors of stroke in groups of patients undergoing cardiac surgery or other invasive manipulations of the aorta, such as cardiac catheterization. Aortic atheroma is a likely site of origin of embolic debris detected in our patients. Our own data suggests that microemboli, ultrasonically detected in the carotid artery during surgery, are the best predictors of a negative neurobehavioral outcome. It is our contention that the prevention of the occurrence and delivery of microemboli is the first line of defense in protecting the brain. This randomized trial will study whether patients who undergo reduced manipulation of the aorta, by means of a single crossclamp and retrograde cardioplegia, have fewer emboli delivered to the brain during cardiac surgery, and a lower combined incidence of neurobehavioral dysfunction following CABG, when compared to a control group of patients where traditional multi-aortic crossclamps are performed in combination with antegrade cardioplegia. The importance of this study is that a positive outcome could result in a very rapid change in surgical procedure that would have an immediate impact on all patients undergoing cardiopulmonary bypass, at no increase in risk or cost to the patient.